Transcranial sonography of brain tumors in the adult: an in vitro and in vivo study.

Abstract

Few reports indicate the potential of transcranial sonography (TCS) in detecting human brain tumors. With an Agilent Sonos 2500 ultrasound device, the authors studied 4 brain tumor phantoms and compared the findings with magnetic resonance imaging (MRI). TCS was performed on 40 patients with intracranial tumors in a follow-up design. Sonographic tumor volume and affection of the ventricular system were compared with MRI findings. The authors found a good correlation between TCS and MRI volumetry in the in vitro study. TCS showed good intraobserver and interobserver reliability. A new volumetric formula for TCS measurement was determined. TCS detection rate of brain tumors in vivo was 40%. When the investigators were given access to radiological findings, the rate of tumor identification was 80%. Despite a sufficient acoustic window, 40% of gliomas grade II and III were not detected. One glioblastoma was not identified owing to an insufficient temporal acoustic window. Tumor volumes measured with MRI and TCS correlated well. MRI volumes exceeded TCS volumes by 41%. In the postoperative examinations (mean = 8 days postoperative, n = 15), the resection cavity was displayed as hyperechogenic. It appeared impossible to differentiate between residual tumor tissue and normal repair mechanisms or blood. In the follow-up examination (mean = 99 days postoperative, n = 15) in 5 patients, neither MRI nor TCS showed tumor regrowth. Ten patients had residual tumors that were detected by sonography. The value of TCS for the diagnostics of brain tumors is at present limited. Once the tumor has been identified, sonographic results match well with those of MRI.